Open
Bug 1523916
Opened 5 years ago
Updated 2 years ago
[socket-proc] Implement backpressure for socket-child data bridge
Categories
(Core :: Networking, enhancement, P3)
Tracking
()
NEW
People
(Reporter: valentin, Unassigned)
References
(Blocks 1 open bug)
Details
(Whiteboard: [necko-triaged])
Attachments
(1 file)
In Bug 1510691 we added the ability to send data directly from the socket process to the child process via PBackground. We to make sure data doesn't pile up in the content process.
Updated•5 years ago
|
Whiteboard: [necko-triaged]
Comment 2•5 years ago
|
||
Yes, the bottom line is to make [1] disappear
https://hg.mozilla.org/projects/larch/rev/3395adff5c4dce4d305407dcff3b73a2b4637249#l20.33
Flags: needinfo?(juhsu)
Updated•5 years ago
|
Assignee: nobody → juhsu
Comment 3•5 years ago
|
||
I think it's okay to keep SendBytesRead even if the ODA is through
Socket-Child. In this case, no flow control in socket process. Doesn't
matter a lot to send some useless BytesRead to parent (it's a no-op,
adding some bytes, not hitting assertion/suspension)
Updated•5 years ago
|
Attachment #9059163 -
Attachment description: Bug 1523916 - Allow Parent-Child bp flow control → Bug 1523916 - Part 1: Allow Parent-Child bp flow control
Comment 4•5 years ago
|
||
https://hg.mozilla.org/projects/larch/rev/d3048bca60d89203e119eb5ddcc1049159543f35
leave open intentionally
Updated•5 years ago
|
Updated•4 years ago
|
Assignee: CuveeHsu → nobody
Updated•2 years ago
|
Severity: normal → S3
You need to log in
before you can comment on or make changes to this bug.
Description
•